JP2006074402A - Imaging apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To widen the range where blurring can be corrected strongly even when the correctable region is constant, e.g. the correction range is not widened by an electronic zoom or the correction range is determined by a structural factor, by having a user select the extent of correction. <P>SOLUTION: The imaging apparatus comprises means 6 and 7 for detecting the blurring oscillation of an imaging element 3 in an optical system 1 in the form of a blurring signal, and a section 12 for operating the correction amount in blurring correction processing for blurring amount data being created based on the blurring signal detected by the blurring detection means depending on zoom position information of the optical system 1, blurring correction type information indicative of the strength of blurring correction processing, distance information from a current correction position to the end of a correctable range, and correction strengthening flag information. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an imaging apparatus having a function of correcting camera shake based on a detected amount of camera shake.

  In recent years, with the downsizing and higher magnification of imaging devices such as video cameras, image blurring and blurring due to the effects of camera shake have become prominent, and in order to solve this problem, imaging devices equipped with a camera shake correction function have been developed. It is getting more.

  There are two methods for detecting the amount of vibration due to camera shake (hereinafter referred to as the amount of camera shake): a method in which the vibration of the imaging unit (device itself) due to camera shake is detected by a sensor (hereinafter referred to as the sensor method), and camera shake based on the captured image. There is a method for detecting the quantity (hereinafter referred to as motion vector method).

  In the sensor system, a detector such as an angular velocity sensor that applies Coriolis force is incorporated into the device, and the amount of vibration of the imaging unit (device itself) due to camera shake is detected. Since it is detected in real time, there is a feature that it is not affected by the state of the subject (type of subject, brightness, movement) when detecting the amount of camera shake.

  On the other hand, in the motion vector method, an image of a subject (light signal) projected through an imaging lens is detected by an imaging device, stored as an image signal in a memory, and compared with an image signal to be detected next. Since image blur is detected as motion vector data, that is, the difference between the two most recent image signals is detected as motion vector data, it affects the state of the optical system such as changes in shooting magnification when detecting the amount of camera shake. There is a feature of not receiving.

Many methods and apparatuses for correcting camera shake based on the above-described sensor and the amount of camera shake detected from a captured image signal have been devised.
There are two types of camera shake correction methods: “optical correction” and “electronic correction”. The former “optical correction” bends the light beam by moving a part of the lens optical system, or by attaching a variable apex angle prism, a rotation / translation (eccentric) lens, etc. to a part of the lens optical system. Thus, camera shake correction is realized.

  The latter “electronic correction” realizes correction by taking out (cutting out) an image so that the amount of camera shake is canceled when the exposed image is read out from the image sensor or after reading out.

  Furthermore, along with miniaturization of cameras, optical systems such as lenses have also been miniaturized. For this reason, in the case where lenses with different magnifications are attached, there are increasing restrictions on the viewable angle range (magnification range).

  In the conventional camera shake correction, in order to remove the noise component of the information of the camera shake detection output obtained by the camera shake detection sensor such as the angular velocity sensor, the filter unit (HPF (HPF (HPF)) is converted to digital data by an AD (Analog to Digital) converter. High Pass Filter), LPF (Low Pass Filter), BPF (Band Pass Filter), etc.) are converted into physical quantity data necessary for camera shake correction.

  For example, when the camera shake detection sensor is an angular velocity sensor, the integration unit performs an integration calculation, and the sensitivity adjustment / zoom adjustment / characteristic compensation unit multiplies a necessary coefficient. Then, correction is performed using one or both of “optical correction device” and “electronic correction” from the signal generator. In the case of optical correction by an optical correction device, the amount of decentering of a lens (shift lens) or the angle of correction by a variable apex angle prism (VAP (Variable Angular Prism)), or CCD (Charge Coupled Device) In the case of electronic correction by reading (cutting out) means, conversion is performed in pixel units such as an image reading position on the image sensor. In the case of an acceleration sensor, integration is performed twice and conversion is performed in the same manner. Thereafter, the correction means is driven to perform actual correction.

  Here, the gain multiplication processing unit of the sensitivity adjustment / zoom adjustment / characteristic compensation unit multiplies the sensitivity adjustment of the sensor and the gain (magnification gain, zoom gain) corresponding to the zoom (focal length). The magnification gain can be switched by preparing means for recognizing the type of lens attached to the camera, and selecting a gain (zoom gain) according to the lens by the gain changing / selecting unit.

  The above-described conventional camera shake correction has only a switch ON / OFF setting indicating whether or not to perform camera shake correction, and the effectiveness of the correction varies depending on the shooting application and the photographer's preference. Therefore, it was not possible to select the degree of correction.

Further, as disclosed in Patent Document 1, the photographer sets an arbitrary value to at least one of the correction ratio between the shake amount of the captured image caused by the shake and the image correction amount at the time of shake correction and the upper limit value of the image correction amount. Although there was a method of “changing the setting”, only the correction range was changed and the correction ratio was changed.
Further, in Patent Document 2, the effectiveness of correction is switched by switching the integral filter frequency characteristics in accordance with switching between the moving image / still image mode. For example, as shown in FIGS. 4 and 8, the frequency range having the integration effect is switched to vary the range in which the correction is effective.

In Patent Document 3, the integration coefficient table of FIG. 10 is changed according to the electronic zoom (as the area that can be corrected increases). When an integration coefficient is selected as in M3 in FIG. 10, it is possible to expand the range in which the correction is strongly applied (the range where the coefficient is close to 1) compared to M1 and M2.
Japanese Patent No. 3243625 Japanese Patent Laid-Open No. 11-308523 Japanese Patent No. 3380918 (FIG. 10)

  However, when the correction range is not widened by the electronic zoom and when the correction range is determined by mechanical structural factors in the optical camera shake correction, the correctable region is constant, so that M3 in FIG. It was not possible to widen the range in which the correction was strongly applied.

  This is because, when the correctable range is constant, if only the integral coefficient is brought close to 1, the end of the correctable range will be reached immediately, the correction will not be effective, or the end will be reached suddenly. Causes an unnatural (discontinuous) movement of the image. Further, when an optical correction unit is used, it may cause noise and vibration. When shooting movies, discontinuities and noise are particularly harmful.

  It should be noted that the case where the correction range is limited to avoid aberrations depending on the position of the stop, zoom, and focus is excluded because the above point is not considered.

  Therefore, according to the present invention, in the camera shake correction, the user selects the strength of the correction, and when the correctable area is constant, for example, when the correction range is not widened by the electronic zoom, and the correction range is obtained by the optical camera shake correction. Even when it is determined by structural factors, it is an object to create a state where a strong correction range is wide.

  In addition, the correction for small amplitude blurring, for example, when the correction lens is near the center of the correction range, while maintaining the effect of correction, weakens the correction for large amplitude, for example, when the correction lens moves away from the center of the correction range. For example, it is possible to reduce a sense of incongruity caused by correcting camerawork at the boundary between intentional motion and camera shake.

  In order to solve the above problems and achieve the object of the present invention, an imaging apparatus of the present invention includes a camera shake detection unit that detects camera shake vibration of an imaging unit of a camera unit as a camera shake signal, and a camera shake signal detected by the camera shake detection unit. Based on the camera shake amount data generated based on the zoom position information of the optical system mounted on the camera unit, camera shake correction type information indicating the type of effect of camera shake correction processing, the distance between the current correction position and the end of the correctable range Correction amount calculation means for calculating a correction amount of camera shake correction processing corresponding to the relationship information and the correction effect change flag information.

  Still further, the imaging apparatus of the present invention includes a pan / tilt detection unit that detects panning and tilting of the camera unit, and a correction amount in the correction amount calculation unit when panning and tilting of the camera unit is detected by the pan / tilt detection unit. A correction unit that limits the correction amount of camera shake correction processing according to distance relationship information from the current correction position to the end of the correctable range for the shake amount data for which the correction amount is calculated by the change unit that changes and the correction amount calculation unit Correction amount changing means for changing the correction effect in accordance with the correction effect change flag information when the panning and tilting of the camera unit is detected by the amount limiting means, the pan / tilt detection means, and the correction amount restriction Image stabilization based on the correction amount limited by the means or the correction amount strengthened by the correction effect changing means In which and a camera-shake correction means for performing a sense.

  As a result, camera shake detection means detects camera shake vibration of the imaging means of the camera unit as a camera shake signal, and generates camera shake amount data based on the detected camera shake signal. Here, with respect to camera shake amount data generated by the correction amount calculation means, zoom position information of the optical system mounted on the camera unit, camera shake correction type information indicating the type of effect of camera shake correction processing, current correction position and correction The correction amount of the camera shake correction process is calculated corresponding to the distance relationship information to the end of the possible range and the correction effect change flag information.

  The panning / tilting of the camera unit is detected by the pan / tilt detection means, and when the panning / tilting of the camera unit is detected by the pan / tilt detection means by the gain data changing means, the correction amount in the correction amount calculating means is changed. Here, for the camera shake amount data for which the correction amount is calculated by the correction amount limiter by the correction amount limiter, the correction amount of the camera shake correction process is limited according to the distance relationship information from the current correction position to the end of the correctable range. To do. At this time, when the panning / tilting of the camera unit is detected by the pan / tilt detection means in the correction strengthening means, the correction effect is changed in accordance with the correction effect change flag information. Therefore, the camera shake correction process of the image signal is executed based on the correction amount limited by the correction amount limiting unit by the correction unit or changed by the correction amount changing unit.

  According to the present invention, in the camera capable of selecting the camera shake correction type, the integration coefficient table is switched according to the user selection by the camera shake correction type information indicating the type of effect of the camera shake correction processing. The trouble that occurs during panning and tilting of the camera unit is automatically detected in the integration coefficient table according to the zoom position information, the current correction means (for example, movement of the lens) and the distance relationship between the correction range and the camera shake correction type information. Mitigates by switching. Further, the frequency characteristic is changed by lowering the pan / tilt detection threshold to make the detection more sensitive or by cutting off the low frequency region.

  In this way, the integration coefficient table is not only changed according to the electronic zoom, but is also changed according to the correction type mode selection by the user, the optical zoom position, and the current position with respect to the limit value of the correction means. The pan / tilt detection is also changed according to the selected correction type mode and the integration coefficient table.

  In addition, since the correction range is limited, if the correction amount reaches the limit, the correction is not effective. At this time, when panning or tilting of the camera unit is detected, the correction amount is increased according to the correction effect change flag information.

  According to the present invention, in a camera with a camera shake correction function, the user selects the strength of correction, and when the correctable area is constant, for example, when the correction range is not widened by electronic zoom, and with optical camera shake correction. Even when the correction range is determined by structural factors, it is possible to create a state in which the range in which the correction is strongly effective is widened.

  In addition, the correction for small amplitude blurring, for example, when the correction lens is near the center of the correction range, while maintaining the effect of correction, weakens the correction for large amplitude, for example, when the correction lens moves away from the center of the correction range. For example, it is possible to reduce a sense of incongruity caused by correcting camerawork at the boundary between intentional motion and camera shake.

First, an imaging apparatus to which an embodiment of the present invention is applied will be described with reference to FIG.
FIG. 1 is a block diagram illustrating a simplified main part of an imaging apparatus to which an embodiment of the present invention is applied. An optical system 1 including an optical correction unit 2 and an imaging element 3, a “camera shake detection unit”. As an example, shake detection means 7 such as an angular velocity sensor, shake detection means 6 such as motion vector detection, signal processing unit 4, camera control microcomputer (microcomputer) 8, hand shake correction block 9, optical correction means drive device 16 and the image sensor driving device 17 and the like.

  First, the camera shake correction block 9 of the camera control microcomputer 8 uses camera shake data generated based on a vibration amount (hereinafter referred to as a camera shake signal) of the image pickup device 3 (device itself) due to camera shake detected by the shake detection means 6 and 7. To do.

  Next, in the camera shake correction block 9 of the camera control microcomputer 8, the correction amount calculation unit 12 calculates the correction amount based on the correction mode, zoom position, and exposure information selected by the user for the camera shake vibration data 23. . Based on the calculated correction amount, an appropriate gain setting is made when it is necessary to change the gain data.

  The gain data is data for adjusting the correction amount of the camera shake correction process. The gain G corresponds to the focal length (zoom) of the image sensor 3, the gain according to the strength of the correction mode setting, and the correction means. It is a gain according to the correction range.

  For example, the correction amount calculator 12 selects the gain data based on the correction amount when the strength of the correction mode is set and the gain data needs to be changed. This gain data is allocated to the optical correction signal generator 14 and the electronic correction signal generator 15 by the allocation 13 to the driving means.

  Subsequently, the optical correction means driving device 16 and the image sensor driving device 17 generate a control signal based on the gain data sent from the camera control microcomputer 8, and the optical correction means of the optical system 1 based on the control signal. 2, or the image signal capture timing and the image signal cut-out range (address) by the image sensor 3 are controlled. Then, the image signal sent from the signal processing unit 4 is subjected to predetermined camera shake correction processing and sent to the recording device 18 at the next stage.

  In the embodiment of the present invention described below, an example is applied to a case where camera shake detection “detects shake of the imaging apparatus itself” and one or both of “optical correction” and “electronic correction” methods are combined. explain.

  FIG. 2 shows a camera shake correction method by the imaging apparatus according to the embodiment of the present invention. FIG. 2 shows the correction calculation unit 12 in the camera shake correction block 9 of the camera control microcomputer 8 which is the main part of FIG.

  In FIG. 2, in order to remove the noise component of the camera shake detection output information S 1 obtained by the camera shake detection means 7 such as an angular velocity sensor, the AD converter 11 converts the information into digital data, and then the panning and tilting detector 31 performs panning. Then, the tilting is detected, and converted into a physical quantity necessary for camera shake correction via the HPF unit 32 and the integration unit (LPF) 33 of the filter unit.

  Further, in order to make it possible to process the signal S1 of the camera shake detection output obtained by the motion detection means 6 such as motion vector detection, the IF converter 10 converts the information S1 into data in a format that can be signal processed, and then performs panning and chilling. The panning and tilting detection is performed by the tilting detection unit 31 and converted into a physical quantity necessary for camera shake correction via the HPF unit 32 and the integration unit (LPF) 33 of the filter unit.

  The panning / tilting detector 31 is provided to detect a user's intentional operation (panning / tilting). FIG. 4 shows an example of the operation. As shown in the figure, detection is performed based on a time threshold value Tthr at the time of transition from the normal state 41 to the detection state 42, a time threshold value Tthr at the time of transition from the detection state 42 to the normal state 43, and an intensity threshold value Sthr.

  The HPF unit 32 removes the noise component of the camera shake detection output obtained by the camera shake detection units 6 and 7 by the configuration of the delay elements 51 and 54, the multiplication elements 52 and 55, and the addition elements 53 and 56 shown in FIG. It is provided to remove high-frequency components different from camera shake and intentional actions of the photographer (panning / tilting, etc.). When panning / tilting is detected by the panning / tilting detector 31, the low frequency is cut off by the panning / tilting information 37 as indicated by 57 in FIG. 5B.

  The integration unit (LPF) 33 integrates the signals obtained by the camera shake detection units 6 and 7. For example, in the case of an angular velocity sensor, integration is performed by the configuration of the delay element 61, the multiplication element 63, and the addition element 64 shown in FIG. 6A, and the necessary coefficient in the multiplication element 63 is multiplied by the input of information 65 to the table 62. In the case of the correction method using the variable apex angle prism 25 shown in FIG. 3, the angle, the amount of eccentricity in the case of correction using the lens eccentric movement 24, and the output region on the imager 26 of the image sensor 3 in the case of electronic correction shown in FIG. Conversion into 28 pixel units. In the case of an angular acceleration sensor, the integration is performed twice and converted in the same manner.

  Here, the information 65 to be input to the table 62 of the integration unit (LPF) 33 includes electronic zoom magnification, camera shake correction type setting, panning and tilting information, distance relationship between the current lens position and limit value, and correction enhancement flag information. It is.

  As shown in FIG. 6B, in the operation of the table of the integration unit (LPF) 33, when panning / tilting is detected by the panning / tilting detection unit 31, a coefficient M3 indicating a stronger correction by the panning / tilting information 37, standard Of the coefficient M2 indicating the correction of the above and the coefficient M1 indicating the weak correction, the coefficient M2 is set close to the coefficient M2 indicating the standard correction.

  In the detailed setting of the table shown in FIG. 6C, a coefficient M3 ′ indicating a slightly stronger correction is provided between a coefficient M3 indicating a stronger correction and a coefficient M2 indicating a standard correction. Further, a coefficient M1 ′ indicating a slightly weaker correction is provided between the coefficient M2 indicating the standard correction and the coefficient M1 indicating the weaker correction. In addition, these coefficient characteristics may be partially intersected or contacted.

  FIG. 7 is a diagram showing another integration (LPF). FIG. 7A shows a table 68 for setting the integration (LPF) 67 and its frequency characteristic based on the information 69. As shown in FIG. In the frequency characteristics of the filter, when panning and tilting are detected by the panning / tilting detector 31, the frequency characteristics F3 having a wide correction frequency range (expressed as strong) and a standard correction frequency range are obtained by the panning and tilting information 37. Of the frequency characteristic F2 and the frequency characteristic F1 having a narrow frequency range (expressed as weak), the frequency characteristic F2 is brought close to the correction frequency range F2 indicating standard correction.

  In the detailed setting of the frequency characteristic of the integration filter shown in FIG. 7C, the frequency characteristic F3 having a slightly wider correction frequency range between the frequency characteristic F3 having a wide correction frequency range and the frequency characteristic F2 having a standard correction frequency range. `Is provided. Further, a frequency characteristic F1 ′ having a somewhat narrow correction frequency range is provided between the frequency characteristic F2 having a standard correction frequency range and the frequency characteristic F1 having a narrow correction frequency range.

  The gain multiplication processing unit of the sensitivity adjustment / zoom adjustment / characteristic compensation unit 35 multiplies the sensitivity adjustment of the sensor and a gain (magnification gain) corresponding to the zoom (focal length).

  Here, the limit value determination unit 34 determines whether or not the distance from the current correction position to the end of the correctable range is a limit value. The strong flag generation unit 36 changes the detection threshold of the panning / tilting detection unit 31 so that the correction amount becomes strong, changes the coefficient settings of the multiplication elements 52 and 55 of the HPF unit 32, and integrates the integration unit ( The coefficient setting of the multiplication element 63 of (LPF) 33 is changed.

  In the case of electronic correction, the correction means sends an image read (cutout) position instruction 20 and an image read (cutout) position instruction 19 as correction values to the image sensor driving device 17 and the memory controller 5. Cut out. In the case of optical correction, an instruction to the optical system such as the translation lens decentering amount of the correction value and the variable apex angle prism angle is sent to the servo control portion of the optical correction means driving device 16 to drive the optical correction means 2. Do.

  In particular, a correction amount restriction unit (not shown) restricts the instruction value according to the correction amount restriction of the correction means. The correction amount constraint is the cutout surplus range of the image sensor 3 if it is electronic correction by the image sensor driving device 17 and the mechanical movable range of the shift lens and VAP in the optical correction by the optical correction means driving device 16.

  Thereafter, in the case of electronic correction, correction values are sent to the image sensor driving device 17 and the memory controller 5 that control the image sensor, and an image is cut out. In the case of optical correction, a correction value is sent to the optical correction means driving device 16 to drive the optical correction means 2.

An operation relating to camera shake correction by the imaging apparatus configured as described above will be described below.
The basic operation is to switch the effectiveness of correction by switching the integration coefficient table of integration processing (LPF), which was conventionally constant outside the electronic zoom region. In the present embodiment, as described below, the integration unit (LPF) 33 table (M1, M2, M3, M1 ′, M3 ′) shown in FIG. Switch and perform transitions. In both cases, correction with a small amplitude (near the center of the correction range) is the same as the conventional one, but has the following characteristics.

  M1 is a coefficient that is weakly corrected for large vibrations. In order to compensate for this drawback, the following method is used. Specifically, when the zoom lens is on the tele side beyond a certain threshold, the table is brought closer to M2. Thereby, the movement of the image at the time of pan / tilt can be made smooth.

M2 is a coefficient indicating a standard correction with a characteristic between M1 and M3.
M3 is a coefficient indicating a correction when trying to correct a large vibration more strongly. At this time, the movement of the image at the time of pan / tilt is slow. In addition, since the correction range is limited, the correction amount reaches the limit value, and correction tends to be ineffective. In order to compensate for this drawback, the following method is used. Specifically, pan / tilt detection is performed sensitively. When pan / tilt is detected, the table is brought close to M2.

Other operations related to camera shake correction by the imaging apparatus configured as described above will be described below.
The basic operation is to switch the effectiveness of correction by switching the integration coefficient table of integration processing (LPF), which was conventionally constant outside the electronic zoom region. In the present embodiment, as will be described below, the frequency characteristics (F1, F2, F3, F1 ′, F3 ′, etc.) of the integration unit (LPF) 33 shown in FIG. 7B according to the camera shake correction type setting input by the user. And switching by this. In either case, correction for vibrations having a high frequency is the same as the conventional one, but has the following characteristics.

  F1 is a frequency characteristic in which correction for a low frequency is weak or uncorrected. In order to compensate for this drawback, the following method is used. Specifically, when the zoom lens is on the tele side for a certain threshold value or more, the frequency characteristic is brought close to F2. Thereby, the movement of the image at the time of pan / tilt can be made smooth.

F2 is a coefficient indicating a standard correction frequency range with characteristics between F1 and F3.
F3 has a characteristic of correcting even a low frequency. At this time, the movement of the image at the time of pan / tilt is slow. In addition, since the correction range is limited, the correction amount reaches the limit value, and correction tends to be ineffective. In order to compensate for this drawback, the following method is used. Specifically, pan / tilt detection is performed sensitively. When pan / tilt is detected, the table is brought close to F2.

  FIG. 8 shows a flowchart of the panning / tilting detection operation. The period threshold value Tthr setting 71 and the intensity threshold value Sthr setting 72 shown in FIG. 4 are switched according to the correction type (T1, 2, 3) and the correction strong flag. The correction enhancement flag will be described in a later flowchart.

  First, in the process of the period threshold value Tthr setting 71, it is determined whether the correction type is set (step S11). When the correction type is set to T1 (step S11), it is determined whether or not the correction enhancement flag is turned on (step S12). When the correction enhancement flag is on (step S12), the period threshold value Tthr setting is brought close to the T1` setting Tthr1` while being smoothed (step S13). When the correction enhancement flag is off (step S12), the period threshold value Tthr is set to the T1 setting Tthr1 (step S14).

  When the correction type is set to T2 (step S11), the period threshold value Tthr is set to T2 setting Tthr2 (step S15). When the correction type is set to T3 (step S11), the period threshold value Tthr is set to the T3 setting Tthr3 (step S16).

  Next, in the process of the intensity threshold value Sthr setting 72, it is determined whether the correction type is set (step S17). When the correction type is set to T1 (step S17), it is determined whether or not the correction enhancement flag is on (step S18). When the correction strength flag is on (step S18), the strength threshold value Sthr is set close to the T1` setting Sthr1` while being smoothed (step S19). When the correction strength flag is off (step S18), the strength threshold value Sthr is set to the T1 setting Sthr1 (step S20).

  When the correction type is set to T2 (step S17), the intensity threshold value Sthr is set to the T2 setting Sthr2 (step S21). When the correction type is set to T3 (step S17), the intensity threshold value Sthr is set to the T3 setting Sthr3 (step S22).

  Then, it is determined whether or not the intensity | S |> the intensity threshold value Sthr (step S23). When the intensity | S |> the intensity threshold value Sthr (step S23), the counter Cnt-1 is incremented (step S24). When the intensity | S |> the intensity threshold value Sthr is not satisfied (step S23), the counter Cnt-1 and the pan / tilt flag are reset (step S25).

  Further, it is determined whether or not counter Cnt-1> intensity threshold value Sthr (step S26). When counter Cnt-1> intensity threshold value Sthr (step S26), the pan / tilt flag is turned on and the process ends (step S27).

  Here, as indicated by 73, for example, the period threshold value Tthr1 <Tthr1`≈Tthr2 <Tthr3, and the intensity threshold value Sthr1 <Sthr1`≈Sthr2 <Sthr3.

  FIG. 9 shows a flowchart of the operation of the HPF. The cut-off of the HPF is transitioned according to the correction type (T1, 2, 3), pan / tilt flag, difference ΔL between the current position and the movable range limit.

  First, in the process of the HPF calculation setting 81, it is determined whether the correction type is set (step S31). When the correction type setting is T1 (step S31), the HPF calculation setting is set to the T1 cutoff setting C1 (step S32). When the correction type is set to T2 (step S31), the HPF calculation setting is set to the T2 cutoff setting C2 (step S33).

When the correction type is set to T3 (step S31), it is determined whether the pan / tilt flag is turned on, or whether the difference ΔL between the current position and the movable range limit> the movable amount threshold value ΔLthr. (Step S34). When the pan / tilt flag is on, or when the difference ΔL between the current position and the movable range limit> the movable amount threshold value ΔLthr (step S34), the HPF calculation setting is smoothed to the T2 cutoff setting C2. (Step S35). Further, when the pan / tilt flag is not turned on, or when the difference ΔL between the current position and the movable range limit is not greater than the movable amount threshold value ΔLthr (step S34), the HPF calculation setting is set to the T3 cutoff setting. C3 is set (step S36).
Then, HPF calculation processing is executed (step S37).

  FIG. 10 shows a flowchart of the operation of integration (LPF) processing. The integration coefficient table is switched according to the correction type (T1, 2, 3), the pan / tilt flag, the difference ΔL between the current position and the movable range limit, and the correction enhancement flag. Further, in this process, a correction enhancement flag is generated. If the zoom position is equal to or greater than a certain threshold value Zthr, pan / tilt is not performed, and the correction means is near the center of the correctable range (the difference ΔL between the current position and the movable range limit> the movable amount threshold value ΔLthr), the counter Cnt_z is incremented. If the counter exceeds the time threshold value Fthr, the correction enhancement flag is turned ON.

  Note that each transition is smoothed to avoid unnatural image movement due to a sudden change in the computation filter.

  First, in the process of the integration coefficient table setting 91, it is determined whether the correction type is set (step S41). When the correction type is set to T1 (step S41), it is determined whether or not the correction enhancement flag is on (step S42). When the correction enhancement flag is on (step S42), the integration coefficient table setting is brought closer to the T1` setting M1` while being smoothed (step S43). When the correction enhancement flag is off (step S42), the integral coefficient table setting is set to the setting M1 for T1 (step S44).

  When the correction type is set to T2 (step S41), the integration coefficient table setting is set to T2 setting M2 (step S45).

  When the correction type is set to T3 (step S41), it is determined whether the pan / tilt flag is on, or whether the difference ΔL between the current position and the movable range limit> the movable amount threshold value ΔLthr. (Step S46). When the pan / tilt flag is on, or the difference ΔL between the current position and the movable range limit> the movable amount threshold value ΔLthr (step S46), the integration coefficient table setting is smoothed to the setting M3` for T3`. (Step S47). If the pan / tilt flag is not turned on or the difference ΔL between the current position and the movable range limit is not greater than the movable amount threshold ΔLthr (step S46), the integration coefficient table setting is set to the T3 setting M3. (Step S48).

  Next, it is determined whether or not the zoom position Z-pos> the zoom position threshold Zthr or more, the difference ΔL between the current position and the movable range limit> the movable amount threshold ΔLthr, and whether the pan / tilt flag is off (step) S49). When zoom position Z-pos> zoom position threshold value Zthr or more, difference ΔL between current position and movable range limit> movable amount threshold value ΔLthr, and the pan / tilt flag is off (step S49), counter Cnt_z is incremented. (Step S50). When the zoom position Z-pos> the zoom position threshold value Zthr or more, the difference ΔL between the current position and the movable range limit> the movable amount threshold value ΔLthr and the pan / tilt flag is not turned off (step S49), the counter Cnt_z is reset. At the same time, the correction enhancement flag is reset (step S51).

Further, it is determined whether or not the counter Cnt_z> the strong threshold value ΔFthr (step S52). When the counter Cnt_z> the strong threshold ΔFthr (step S52), the correction strong flag is turned on (step S53). When the counter Cnt_z> the strong threshold ΔFthr is not satisfied (step S52), the correction strong flag is reset (step S54).
Then, integration (LPF) calculation processing is executed (step S55).

  Further, FIG. 11 shows that the coefficients M1, M2, M3, M1 ′, M2 ′, and M3 ′ in FIG. 10 are replaced with frequency characteristics F1, F2, F3, F1 ′, F2 ′, and F3 ′, as indicated by 111. The frequency characteristics of the integration filter are variably set and the other operations are the same as those in FIG.

  Needless to say, the present invention is not limited to the above-described embodiment, and other configurations can be appropriately adopted within the scope of the claims of the present invention.

It is a figure which shows the structure of the camera-shake correction by the imaging device of embodiment of this invention. FIG. 2A is a diagram illustrating a camera shake correction method, FIG. 2A is a case of optical correction, and FIG. 2B is a case of electronic correction. It is a block diagram which shows the structure of a correction amount calculating part. It is a figure which shows operation | movement of a panning / tilting detection part. FIG. 5A is a diagram showing the HPF, FIG. 5A is a diagram showing the configuration of the HPF, and FIG. 5B is a diagram showing the operation of the HPF. FIG. 6A is a diagram showing an integration (LPF), FIG. 6A is a configuration of integration (LPF), FIG. 6B is a table operation, and FIG. 6C is a diagram showing detailed table settings. FIG. 7A is a table for setting integration (LPF) and its frequency characteristic, FIG. 7B is a frequency characteristic of the integration filter, and FIG. 7C is a detailed setting of the frequency characteristic of the integration filter. FIG. It is a flowchart which shows the operation | movement of a panning / tilting detection. It is a flowchart which shows operation | movement of HPF. It is a flowchart which shows the operation | movement of integration (LPF). It is a flowchart which shows the operation | movement of another integration (LPF).

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 ... Optical system, 2 ... Optical correction means, 3 ... Image pick-up element, 4 ... Signal processing part, 5 ... Memory controller, 6 ... Shake detection means (motion vector detection), 7 ... Shake detection means (angular velocity sensor), 8 ... Microcomputer for camera control, 9 ... Shake correction block, 10 ... Interface, 11 ... A / D conversion unit, 12 ... Correction amount calculation unit, 13 ... Allocation unit to driving means, 14, 15 ... Signal generator, 16 ... Optical Correction means driving device, 17 ... imaging device driving device, 18 ... recording device, 19, 20 ... image reading (cutting) position instruction, 21 ... instruction to optical system such as translational lens decentering amount, variable apex angle prism angle, 22 ... Use either or both correction methods, 24 ... Lens movement (translation), 25 ... VAP (variable vertical angle prism), 26 ... Imager, 27 ... A point output area, 28 ... B point output area , 31 ... Panning / tilting detection unit, 32 ... HPF, 33 ... Integration (LPF), 34 ... Limit value determination unit, 35 ... Sensitivity adjustment / zoom adjustment / characteristic compensation unit, 36 ... Strong flag generation unit, 37 ... Panning / Tilting information, 38 ... Shake correction type setting, 39 ... Shake correction type setting, zoom position information, relationship between current correction position and limit value

Claims (8)

Camera shake detection means for detecting camera shake vibration of the imaging means of the camera unit as a camera shake signal;
For camera shake data generated based on the camera shake signal detected by the camera shake detection means, zoom position information of the optical system mounted on the camera unit, camera shake correction type information indicating the type of effect of camera shake correction processing, Correction amount calculation means for calculating a correction amount of camera shake correction processing corresponding to the distance relationship information from the current correction position to the end of the correctable range and the correction effect change flag information;
Pan / tilt detection means for detecting panning and tilting of the camera unit;
Changing means for changing the correction amount in the correction amount calculation means when panning and tilting of the camera unit is detected by the pan / tilt detection means;
Correction amount limiting means for limiting the correction amount of the camera shake correction processing according to the distance relation information from the current correction position to the end of the correctable range with respect to the camera shake amount data whose correction amount has been calculated by the correction amount calculating means. When,
Correction effect changing means for changing the correction amount in the limiting means according to the correction effect change flag information when panning and tilting of the camera unit is detected by the pan / tilt detection means;
A camera shake correction unit that performs a camera shake correction process on the image signal based on a correction amount limited by the correction amount limiting unit or a correction amount whose effect is changed by the correction effect changing unit. An imaging device. The imaging device according to claim 1,
The correction amount calculation unit has a correction amount table storing the correction amount, and the correction is performed when panning and tilting of the camera unit is detected by the pan / tilt detection unit and based on the correction effect change flag information. An imaging apparatus, wherein the correction amount is changed by selectively switching an amount table. The imaging device according to claim 2,
A menu setting unit capable of setting the image stabilization type information;
An imaging apparatus, wherein the correction amount is changed by switching the correction amount table based on the camera shake correction type information set by the menu setting means. The imaging device according to claim 2,
The imaging apparatus according to claim 1, wherein the changing unit changes the correction amount by switching the correction amount table so as to block a low frequency region of the camera shake amount data in the correction amount calculating unit. The imaging device according to claim 2,
The imaging apparatus according to claim 1, wherein the changing unit changes the correction amount by switching the correction amount table so as to change an integration coefficient in the correction amount calculating unit. The imaging device according to claim 2,
The imaging apparatus according to claim 1, wherein the changing unit changes the correction amount by switching the correction amount table so as to vary a frequency characteristic of the integral filter in the correction amount calculating unit. The imaging device according to claim 2,
The changing means does not change the correction amount when the current correction position is near the center of the correctable range, and changes the correction amount to be small when the current correction position is near the end of the correctable range. An imaging device that is characterized. The imaging device according to claim 2,
The image pickup apparatus according to claim 1, wherein when the correction amount is changed, the change unit changes the panning / tilting detection threshold of the camera unit by the pan / tilt detection unit.

Images